Am 19.03.2011 16:53, schrieb Bart Coninckx: > Don't your logfiles reveal anything? >
I see the following messages in the logfile of the first node. There is a pause from 21:36:25 to 21:37:21. It says that an Election Trigger popped after 60 seconds. I wonder why a timeout is triggered here. Both nodes work without problems. Mar 19 21:36:22 laplace cib: [23840]: info: ais_dispatch_message: Membership 6820: quorum still lost Mar 19 21:36:22 laplace cib: [23840]: info: crm_new_peer: Node <null> now has id: 33663168 Mar 19 21:36:22 laplace cib: [23840]: info: crm_update_peer: Node (null): id=33663168 state=member (new) addr=r(0) ip(192.168.1.2) votes=0 born=0 seen=6820 proc=00000000000000000000000000000000 Mar 19 21:36:22 laplace cib: [23840]: notice: ais_dispatch_message: Membership 6820: quorum acquired Mar 19 21:36:22 laplace cib: [23840]: info: crm_get_peer: Node 33663168 is now known as ries Mar 19 21:36:22 laplace cib: [23840]: info: crm_update_peer: Node ries: id=33663168 state=member addr=r(0) ip(192.168.1.2) votes=1 (new) born=6820 seen=6820 proc=00000000000000000000000000151312 (new) Mar 19 21:36:22 laplace corosync[23832]: [CPG ] chosen downlist from node r(0) ip(192.168.1.1) Mar 19 21:36:22 laplace corosync[23832]: [MAIN ] Completed service synchronization, ready to provide service. Mar 19 21:36:22 laplace crmd: [23845]: info: te_connect_stonith: Connected Mar 19 21:36:22 laplace crmd: [23845]: info: ais_dispatch_message: Membership 6820: quorum still lost Mar 19 21:36:22 laplace crmd: [23845]: info: crm_new_peer: Node <null> now has id: 33663168 Mar 19 21:36:22 laplace crmd: [23845]: info: crm_update_peer: Node (null): id=33663168 state=member (new) addr=r(0) ip(192.168.1.2) votes=0 born=0 seen=6820 proc=00000000000000000000000000000000 Mar 19 21:36:22 laplace crmd: [23845]: notice: ais_dispatch_message: Membership 6820: quorum acquired Mar 19 21:36:22 laplace crmd: [23845]: info: crm_get_peer: Node 33663168 is now known as ries Mar 19 21:36:22 laplace crmd: [23845]: info: ais_status_callback: status: ries is now member Mar 19 21:36:22 laplace crmd: [23845]: notice: crmd_peer_update: Status update: Client ries/crmd now has status [online] (DC=<null>) Mar 19 21:36:22 laplace crmd: [23845]: info: crm_update_peer: Node ries: id=33663168 state=member addr=r(0) ip(192.168.1.2) votes=1 (new) born=6820 seen=6820 proc=00000000000000000000000000151312 (new) Mar 19 21:36:25 laplace attrd: [23842]: info: cib_connect: Connected to the CIB after 1 signon attempts Mar 19 21:36:25 laplace attrd: [23842]: info: cib_connect: Sending full refresh Mar 19 21:37:21 laplace crmd: [23845]: info: crm_timer_popped: Election Trigger (I_DC_TIMEOUT) just popped! (60000ms) Mar 19 21:37:21 laplace crmd: [23845]: WARN: do_log: FSA: Input I_DC_TIMEOUT from crm_timer_popped() received in state S_PENDING Mar 19 21:37:21 laplace crmd: [23845]: info: do_state_transition: State transition S_PENDING -> S_ELECTION [ input=I_DC_TIMEOUT cause=C_TIMER_POPPED origin=crm_timer_popped ] Mar 19 21:37:21 laplace crmd: [23845]: info: do_state_transition: State transition S_ELECTION -> S_PENDING [ input=I_PENDING cause=C_FSA_INTERNAL origin=do_election_count_vote ] Mar 19 21:37:21 laplace crmd: [23845]: info: do_dc_release: DC role released Mar 19 21:37:21 laplace crmd: [23845]: info: do_te_control: Transitioner is now inactive Mar 19 21:37:21 laplace crmd: [23845]: info: update_dc: Set DC to ries (3.0.5) The second node basically says: Mar 19 21:36:18 ries crmd: [19625]: info: crm_update_peer: Node ries: id=33663168 state=member (new) addr=r(0) ip(192.168.1.2) (new) votes=1 (new) born=6820 seen=6820 proc=00000 000000000000000000000151312 (new) Mar 19 21:36:18 ries crmd: [19625]: info: crm_new_peer: Node laplace now has id: 16885952 Mar 19 21:36:18 ries crmd: [19625]: info: crm_new_peer: Node 16885952 is now known as laplace Mar 19 21:36:18 ries crmd: [19625]: info: ais_status_callback: status: laplace is now unknown Mar 19 21:36:18 ries crmd: [19625]: info: ais_status_callback: status: laplace is now member (was unknown) Mar 19 21:36:18 ries crmd: [19625]: info: crm_update_peer: Node laplace: id=16885952 state=member (new) addr=r(0) ip(192.168.1.1) votes=1 born=6820 seen=6820 proc=00000000000000 000000000000151312 Mar 19 21:36:18 ries crmd: [19625]: info: do_started: Delaying start, Config not read (0000000000000040) Mar 19 21:36:18 ries crmd: Last message '[19625]: info: do_st' repeated 1 times, supressed by syslog-ng on ries.site Mar 19 21:36:18 ries crmd: [19625]: info: config_query_callback: Shutdown escalation occurs after: 1200000ms Mar 19 21:36:18 ries crmd: [19625]: info: config_query_callback: Checking for expired actions every 900000ms Mar 19 21:36:18 ries crmd: [19625]: info: config_query_callback: Sending expected-votes=2 to corosync Mar 19 21:36:18 ries crmd: [19625]: info: do_started: The local CRM is operational Mar 19 21:36:18 ries crmd: [19625]: info: do_state_transition: State transition S_STARTING -> S_PENDING [ input=I_PENDING cause=C_FSA_INTERNAL origin=do_started ] Mar 19 21:36:19 ries crmd: [19625]: info: ais_dispatch_message: Membership 6820: quorum retained Mar 19 21:36:19 ries crmd: [19625]: info: te_connect_stonith: Attempting connection to fencing daemon... Mar 19 21:36:20 ries crmd: [19625]: info: te_connect_stonith: Connected Mar 19 21:36:22 ries attrd: [19623]: info: cib_connect: Connected to the CIB after 1 signon attempts Mar 19 21:36:22 ries attrd: [19623]: info: cib_connect: Sending full refresh Mar 19 21:36:22 ries dhclient: XMT: Solicit on eth0, interval 108990ms. Mar 19 21:37:17 ries crmd: [19625]: info: do_election_count_vote: Election 2 (owner: laplace) pass: vote from laplace (Uptime) Mar 19 21:37:17 ries crmd: [19625]: info: do_state_transition: State transition S_PENDING -> S_ELECTION [ input=I_ELECTION cause=C_FSA_INTERNAL origin=do_election_count_vote ] Mar 19 21:37:17 ries cib: [19621]: info: cib_process_readwrite: We are now in R/W mode Mar 19 21:37:17 ries attrd: [19623]: info: find_hash_entry: Creating hash entry for terminate Mar 19 21:37:17 ries pengine: [19624]: notice: unpack_config: On loss of CCM Quorum: Ignore _______________________________________________ Linux-HA mailing list [email protected] http://lists.linux-ha.org/mailman/listinfo/linux-ha See also: http://linux-ha.org/ReportingProblems
